Last verified
ARCHIVE PRICERETIRED MODEL200KTEXT + VISIONPROMPT CACHING

Claude 3.5 Sonnet v2 API Pricing

Claude 3.5 Sonnet v2 is retired; this archive keeps its $3/$15 API price. $3/M input, $15/M output, and $0.3/M cache-hit input. The October 2024 upgrade kept the same price tier as the June 2024 Sonnet 3.5 release while improving coding and computer-use behavior.

Archived input - per 1M tokens
$3.00/M
Source Anthropic retired
Archived output - per 1M tokens
$15.00/M
Use for invoice checks retired
Cached input - 90% off
$0.30/M
Cache reads -90%
Effective - agentic blend
$2.05/M
92/8 split - 82% cache
§ 01 / TERMINAL

Claude 3.5 Sonnet v2 API cost calculator.

Archive calculator pre-loaded with Claude 3.5 Sonnet v2 rates. Tweak spend, output mix, or cache hit rate to compare historical bills with current replacements.

$ /mo
Workload split
Prompt cache hit rate
Tokens you can process
—
Words equivalent (English)
—
Effective rate
—
Open full calculator (all models · share URL · CSV) →
§ 02 / SCENARIOS

Claude 3.5 Sonnet v2 cost on real workloads.

CODING AGENT RETIRED

Repo-wide bug fix

$0.180/task
81k in - 7k out~556 units/$100
LONG DOC ANALYSIS RETIRED

Reading 100-page contracts

$0.281/doc
175k in - 8k out~355 units/$100
RAG SUPPORT RETIRED

Support agent ticket triage

$0.019/ticket
4k in - 1k out~5,352 units/$100
ASSISTANT TURN RETIRED

Research planning turn

$0.041/turn
12k in - 2k out~2,435 units/$100
§ 03 / TOKENIZER

Claude 3.5 Sonnet v2 token counter and cost.

What's a token →
Estimate · anthropic-bpe-estimate · ≈3.5 chars/token Auto-counts as you type

This is a chars-per-token approximation, not a real tokenizer. Actual tokens vary by language, code density, and tool-call overhead — counts are typically ±10–20% off for English prose, more for code or non-Latin scripts. For exact billing, use the vendor's official tokenizer.

Characters 317
Words 48
Tokens (estimated) 91 tokens
Cost as input · uncached $0.00027 USD
Cost as output · uncached $0.00136 USD
Cost as cached input $0.00003 USD
§ 04 / SHELF

Claude 3.5 Sonnet v2 vs similar models.

All models →
Model Input /M Output /M Effective blended Context Best for
Claude Sonnet 3.5 (October 2024) Current $3.00 cache $0.30 $15.00 $2.05 agentic 92/8 200K Production Claude agents
Claude Sonnet 4.6 $3.00 cache $0.30 $15.00 $2.05 same 1M Production Claude agents
Claude Sonnet 4.5 $3.00 cache $0.30 $15.00 $2.05 deprecated · same 200K Production Claude agents
Claude Haiku 4.5 $1.00 cache $0.10 $5.00 $0.682 −3.0× 200K Low-latency Claude workloads
Claude Opus 4.7 $5.00 cache $0.50 $25.00 $3.41 +67% 1M Frontier reasoning and hard code
GPT-6.1 Sol $2.00 cache $0.10 $10.00 $1.29 −37% 1.05M OpenAI's balanced model for coding and agents
Gemini 3.8 Flash $0.75 cache $0.075 $3.75 $0.481 −4.3× · promo 1M Gemini long-context work
DeepSeek V4 Pro $1.32 cache $0.044 $3.96 $0.569 −3.6× 1M Low-cost reasoning and coding
§ 05 / DEEP LINKS

Claude 3.5 Sonnet v2 cost by scenario.

All calculators →

By migration question

Compare against Claude Sonnet 4.6 → Compare against Claude Haiku 4.5 → Compare against Claude Opus 4.7 →

Claude 3.5 Sonnet v2 API pricing FAQ.

Short answers for teams checking old Claude 3.5 Sonnet v2 costs or migrating archive workloads.

Q · 01 Is Claude 3.5 Sonnet v2 still available? +
Claude 3.5 Sonnet v2 is retired on the first-party Claude API. Keep this page for archive pricing, old invoices, and migration planning.
Q · 02 What replaced Claude 3.5 Sonnet v2? +
Anthropic lists claude-sonnet-4-6 as the recommended replacement in its model deprecations page.
Q · 03 How much did Claude 3.5 Sonnet v2 cost? +
Claude 3.5 Sonnet v2 is archived at $3/M input and $15/M output.
Q · 04 Does this archive include prompt caching? +
Cache-hit input is archived at $0.3/M, matching Anthropic's 10% cache-read multiplier on this price tier.
Q · 05 Can I use this for new production routing? +
Use the archive for cost history, not fresh routing decisions. Compare the current Claude rows in the shelf before starting new workloads.
Q · 06 Why keep a page for a retired model? +
Retired-model pages help teams reconcile historical spend, understand migration economics, and answer old search demand without pretending the model is current.
Y
Reviewed by Yaroslav Vikhariev Founder · AI//COST · Prices checked by hand against anthropic.com · Last verified May 18, 2026
Methodology Report a correction More by Y.V.